Abstract
AbstractWe construct a class of non-reversible Metropolis kernels as a multivariate extension of the guided-walk kernel proposed by Gustafson (Statist. Comput.8, 1998). The main idea of our method is to introduce a projection that maps a state space to a totally ordered group. By using Haar measure, we construct a novel Markov kernel termed the Haar mixture kernel, which is of interest in its own right. This is achieved by inducing a topological structure to the totally ordered group. Our proposed method, the$\Delta$-guided Metropolis–Haar kernel, is constructed by using the Haar mixture kernel as a proposal kernel. The proposed non-reversible kernel is at least 10 times better than the random-walk Metropolis kernel and Hamiltonian Monte Carlo kernel for the logistic regression and a discretely observed stochastic process in terms of effective sample size per second.
Publisher
Cambridge University Press (CUP)
Subject
Statistics, Probability and Uncertainty,General Mathematics,Statistics and Probability
Reference66 articles.
1. Bayes and big data: the consensus Monte Carlo algorithm;Scott;Internat. J. Manag. Sci. Eng. Manag.,2016
2. Geometric;Roberts;J. Appl. Prob.,2001
3. Optimal scaling of discrete approximations to Langevin diffusions;Roberts;J. R. Statist. Soc. B [Statist. Methodology],1998
4. Geometric ergodicity and hybrid Markov chains;Roberts;Electron. Commun. Prob.,1997
5. Central limit theorem for additive functionals of reversible Markov processes and applications to simple exclusions;Kipnis;Commun. Math. Phys.,1986